1. Field of the Invention
The present invention relates to a failure recovery system and a server. More particularly, it relates to a failure recovery system and a server in a network control in which, from among countermeasure candidates against failures as have been accumulated in a network operation management, an effective countermeasure is selected on the basis of priority degrees learnt automatically, thereby to recover from a failure.
2. Description of the Related Art
In a conventional network management, a countermeasure method in the case where a failure has occurred in a network has been such that a highly skilled operator directly deals with each failure, or that operating methods for recovering the network from simple failures are prepared as a script on a computer beforehand, whereupon an operator selects the corresponding operating method. In the case of the former method, however, the countermeasure method depends upon the skill of the operator, so that a standardized operation is difficult to be performed, and the human cost of the operator himself/herself tends to become very high. Besides, in the case of the latter, the operating methods for eliminating the failures can be held beforehand, and hence, the operator is relieved from the necessity of dealing with the failures in failure units. However, which of the failures has occurred needs to be specified and judged, and a complete automation has not been made.
Besides, a failure information management scheme in a network supervision system has been disclosed (refer to, for example, Japanese Patent No. 3,099,770 and Japanese Patent No. 3,618,682). This scheme includes a failure information management table in which failure sorts that might occur, influence priority degrees that are previously determined in correspondence with the respective failure sorts and in accordance with the magnitudes of influence degrees exerted by the corresponding failures, and influence failure sorts that might be exerted by the corresponding failures are predetermined and stored. Besides, in response to the occurrences of a plurality of failures, the failure sort forming the cause of the failures is specified by reference to the failure information management table, and in accordance with the influence priority degrees and the influence failure sorts which correspond to the respective failure sorts having occurred.